Academic Journals Database
Disseminating quality controlled scientific knowledge

A New Classification Approach for PolSAR Images

ADD TO MY LIST
 
Author(s): T. J. Singh Chugh | R. B. Dubey | Sujata

Journal: International Journal of Advanced Computer Research
ISSN 2249-7277

Volume: 2;
Issue: 5;
Start page: 170;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Feature evaluation | feature selection | multifeature combination | SAR image classification

ABSTRACT
PolSAR images provide more information than theconventional radar images and so greatly improves theability to terrain discrimination. This work is focused onthe problem of the classification of different terrain typesof PolSAR images. The proposed method uses acombination of two different classifiers. This method isnamed as Meta classifier and compared with othercompetitive classifiers. One as a spatial partitioningmethod, assigning each leaf of tree a distinct region label.So they are called as clustering trees. The other one is usedas the second classifier for generating the class labels. Butthe final result is generated by voting over the ensembleresults of the two classifiers. The experimental resultsobtained so far, indicate the potential of this approach.
Save time & money - Smart Internet Solutions      Why do you need a reservation system?